calufa

LLMs are hungry for tokens. Every day I see more startups claiming token usage at 50-100k per month. You can always brute-force your way in - just see HuggingFace's recent attacks.

If throwing more money at inference while accumulating compounding technical debt is the new norm, then we are not solving the problem, and the solution space is already covered.

Perhaps there are marginal gains at the expense of quadrillion-params LLM models with 10x the cost and energy. We are simply making inefficiency more expensive, camouflaged by VC money and great marketing.

If that is not plateauing, then I guess I will have to reconsider what plateauing means.

As LLM coding agents plateau— at least for the average engineer without tens of thousands of dollars or swarms of agents to run —I’d say that, from here on it’s going to be about ASICs, specialized LoRA/or-equivalent models, or a Ruby on Rails for LLM context engineering and orchestration, which LangChain and others seems well position, including Google as they own the entire stack. LLM free lunch has been over for a while, perhaps since the ReAct loop, and has been official since Ilya mentioned it at NeurIPS.

I feel the most exciting development these days is self-evolving agents. Especially if you have a way to verify their outputs with a formal system, or with a system developed since the 60s by armies of PhDs.

DeepMinds Gnome is a good example, where they use DFT to verify outputs. Approximating NP-problems is always fun for those who dare.

I am also building in this space. Its a mix between HPC, AI, and hard science. Pretty fun compared to waking everyday to LLM news that seem more like marketing stunts.

Better technology cannot compensate for poor product design. Two weeks' worth of tokens sounds like tens of thousands of dollars by the time it is done. Wouldn't it be better to hire someone who knows what they are doing, get it right, and teach the other engineers why and how? Tech dept and product dept creeping in on every LLM loop, compounding.

Author here. High-level:

- Problem: AI UI generators are high-fidelity by default → teams bikeshed aesthetics before structure is right.

- Idea: use ASCII as an intentionally low-fidelity “layout spec” to lock hierarchy/flow first.

Why ASCII: - forces abstraction (no colors/fonts/shadows)

- very fast to iterate (seconds)

- pasteable anywhere (Slack/Notion/GitHub)

- editable by anyone

Workflow:

- describe UI → generate ASCII → iterate on structure/states → feed into v0/Lovable/Bolt/etc → polish visuals last

It also facilitates discussion:

- everyone argues about structure/decisions, not pixels

- feedback is concrete (“move this”, “add a section”), not subjective

More advanced setups could integrate user/customer support feedback to automatically propose changes to a spec or PRD, enabling downstream tasks to later produce PRs.

clarity, purpose and a path for career growth

Just a heads up, but I tried to signup but the button doesn't seem to work.